[Eeglablist] Problems with scripting

arno arno at salk.edu
Fri Jan 26 08:12:16 PST 2007


Dear Michael,

thank you for your email. About the graphic crashes, I do not really 
know. We have Opterons 64-bits and we are running I think Fedora (Kernel 
is 2.6.9-22.EL) and Matlab SP3 and it works fine.
About your scripting problem. There is indeed sometimes a problem with 
the pop_chanedit function about importing channel location files from 
the command line (it is usually about looking some electrode location up 
not about loading them though). I have attached the updated function. 
The referencing problem might be a consequence of that too? Please let 
us know at eeglab at sccn.ucsd.edu if it works.

Best regards,

Arno

wibral at mpih-frankfurt.mpg.de wrote:
> Dear Listusers,
>
> I have the following problemes with EEGLAB (or rather it's Matlab's
> fault but I'd be interested in a workaround). I am using MATLAB R2006b
> on Red Hat Enterprise Linux 64 bit. The GUI works stable for a short
> while, but repeated calls to diffrent datasets or a couple of ICA
> decompositions make it fail (as do many other leghtish calculations that
> are not part of EEGLAB). This gets better (of course) if I start without
> the desktop. But to avoid the chrashes I really need to start without
> the jvm.
> I therefore tried to do everything by scripting. However, the following
> code failed to  save the channel  location and rereferencing
> information, but curiously enough it saved the ICA weights, any ideas?
> (b.t.w. I can't use the standard [ALLEEG EEG Currentset]=eeglab; as I
> immedeatly get a crash because of the call to the gui there...)
>
> %%%%%%%%%%%%%%%%%%%CODE%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> %clear all;
> SubjectFileNames={
>     'jn_wm_p3.vhdr'
>     'kb_wm_p3_2.vhdr' %eeg data missing!
>     'mc_wm_p3.vhdr'
>     'mm_wmp3_140304.vhdr'
>     'ns_wm_p3.vhdr'
>     'pu_wm_p3.vhdr'
>     'rc_wm_p3.vhdr'
>     'ta_mw_p3_64e.vhdr'
>     'vv_wm_p3.vhdr'
>     };
> ELPFileNames={
>     'jn_wm_p3.elp'
>     'kb_wm_p3_2.elp'
>     'mc_wm_p3.elp'
>     'mm_wmp3_140304.elp'
>     'ns_wm_p3.elp'
>     'pu_wm_p3.elp'
>     'rc_wm_p3.elp'
>     'ta_mw_p3_64e.elp'
>     'vv_wm_p3.elp'
>     }
>
> if ~(length(SubjectFileNames)==length(ELPFileNames))
>     disp('Non matching number of header and electrodes files!
> Exiting!');
>     return
> end
>
> for n=1:length(SubjectFileNames)
>     dispStr=strcat('Processing subject #: ', num2str(n));
>   sp(dispStr)
>     %%%%%%%%%%%%%%%%%%%%%%%%EEGLAB part
>    if n>1
> clear EEG;
> end
>  
> EEG = pop_loadbv('/home/wibral/Projects/ICADIPOLE/raw_63/',
> SubjectFileNames{n}, 1, [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
> 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
> 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63]);
> EEG.setname=SubjectFileNames{n};
> EEG=pop_chanedit(EEG,  'load',{
> strcat('/home/wibral/Projects/ICADIPOLE/raw_63/',ELPFileNames{n}),...
>     'filetype', 'besa'});
> EEG = pop_select( EEG, 'channel',[2:63] ); % Do not use data from  
> %channel 1 = 'VEOG' and 64='REF' (empty)
> EEG = reref( EEG, [], 'refstate','common');
> EEG=pop_runica(EEG, 'icatype','fastica','approach','symm', 'g', 'tanh')
> EEG = eeg_checkset( EEG);
> NewFileName=strcat(SubjectFileNames{n},'_EEGonly_RR_ICA.set');
> EEG.setname=NewFileName; 
> save(NewFileName, 'EEG');
> end
> %%%%%%%%%%%%%%%END of CODE%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> I would also appreciate any suggestions about the best Linux 64 bit
> distribution to run MATLAB R2006b, as RHEL4/CentOS4 is a real nightmare
> (countless jvm and glibc issues).
>
> Michael Wibral
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>
>
>   

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pop_chanedit.m
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20070126/6439f70e/attachment.ksh>


More information about the eeglablist mailing list